\begin{table}[h] \centering
\newcolumntype{R}{>{\raggedleft\arraybackslash}X}
\newcolumntype{L}{>{\raggedright\arraybackslash}X}
\newcolumntype{C}{>{\centering\arraybackslash}X}

\caption{Effects of turnovers in the executive branch on country performance}
\label{tab:alternative_sharp_fuzzy}
\begin{tabularx}{\linewidth}{@{}lCCCCCCCC@{}}

\toprule
& (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8) \tabularnewline 
{}&{\textbf{Econ. perf.}}&{GDP p.c. gr.}&{(Minus) Inflation}&{(Minus) Unemp.}&{Trade}&{\textbf{HDI}}&{\textbf{Democ.}}&{\textbf{General index}} \tabularnewline
\midrule \addlinespace[\belowrulesep]
\multicolumn{9}{l}{\textbf{Panel a: Fuzzy RDD (Effects of an executive turnover)}} \\ \hline Ex. turn.&0.367***&--0.026&0.398*&0.811***&0.494***&0.521**&0.058&0.343*** \tabularnewline
&(0.161)&(0.212)&(0.252)&(0.333)&(0.196)&(0.260)&(0.160)&(0.154) \tabularnewline
p-val.&[0.005]&[0.977]&[0.055]&[0.005]&[0.003]&[0.023]&[0.868]&[0.009] \tabularnewline
\hline N&1582&1288&1378&962&1240&940&1529&1677 \tabularnewline
N eff.&674&689&676&444&553&459&831&723 \tabularnewline
Band.&16.7&23.1&19.7&18.8&16.9&20.5&22.6&17.2 \tabularnewline
\hline \multicolumn{9}{l}{\textbf{Panel b: Reduced form (Effects of a defeat of the leading party before the election)}} \\ \hline El. defeat&0.274***&--0.015&0.315*&0.594***&0.354***&0.361**&0.027&0.256** \tabularnewline
&(0.118)&(0.166)&(0.188)&(0.235)&(0.142)&(0.187)&(0.109)&(0.113) \tabularnewline
p-val.&[0.007]&[0.986]&[0.051]&[0.004]&[0.004]&[0.034]&[0.960]&[0.011] \tabularnewline
\hline N&1582&1288&1378&962&1240&940&1529&1677 \tabularnewline
N eff.&571&628&576&385&517&429&739&635 \tabularnewline
Band.&13.4&20.9&15.9&15.5&15.3&18.5&19.8&14.3 \tabularnewline
\hline \multicolumn{9}{l}{\textbf{Panel c: First stage (Effects of a defeat of the leading party before the election on the probability of an}} \\ \multicolumn{9}{l}{\textbf{executive turnover)}} \\ \hline El. defeat&0.697***&0.745***&0.685***&0.691***&0.678***&0.687***&0.660***&0.684*** \tabularnewline
&(0.068)&(0.062)&(0.069)&(0.077)&(0.075)&(0.082)&(0.064)&(0.067) \tabularnewline
p-val.&[$<$0.001]&[$<$0.001]&[$<$0.001]&[$<$0.001]&[$<$0.001]&[$<$0.001]&[$<$0.001]&[$<$0.001] \tabularnewline
\hline N&1582&1288&1378&962&1240&940&1529&1677 \tabularnewline
N eff.&674&689&676&444&553&459&831&723 \tabularnewline
Band.&16.7&23.1&19.7&18.8&16.9&20.5&22.6&17.2 \tabularnewline
\hline \multicolumn{9}{l}{\textbf{Panel d: OLS (Effects of an executive turnover)}} \\  \hline Ex. turn.&0.084***&0.127**&0.071&--0.070&0.155***&--0.010&0.046&0.073** \tabularnewline
&(0.032)&(0.055)&(0.053)&(0.068)&(0.055)&(0.063)&(0.042)&(0.035) \tabularnewline
p-val.&[0.009]&[0.021]&[0.180]&[0.305]&[0.005]&[0.879]&[0.275]&[0.037] \tabularnewline
\hline N&1582&1288&1378&962&1240&940&1529&1677 \tabularnewline
\bottomrule \addlinespace[\belowrulesep]

\end{tabularx}
\\ \parbox{\linewidth}{\footnotesize \caption*{\footnotesize \emph{Notes}: This table reports estimated effects of turnovers in the executive branch for the sample of elections leading to the appointment of a leader in the executive branch. In panel a, we report fuzzy RDD estimates of the effects of executive turnovers, using as assignment variable the defeat of the leading party before the election and turnover in the executive branch as treatment. We show estimates of $\gamma$ in equation (\ref{eq:alternative_rdd}) -- see Appendix \ref{app:executive_turnover}. In panel b, we report reduced form estimates of $\gamma^r$ in equation (\ref{eq:alternative_rdd_rf}) -- see Appendix \ref{app:executive_turnover}, corresponding to the effects of an electoral defeat of the leading party before the election. We use the margin of victory of the best ranked challenger of the leading party before the election as the running variable. In panel c, we report estimates of $\tilde \gamma$ in equation (\ref{eq:alternative_rdd_fs}), corresponding to the first stage of panel a. In panel d, we report OLS estimates of the effect of executive turnovers. Details about the definition of the leading party before the election can be found in Section \ref{sec:turnoverAlt}. $^{*} p<0.10,^{**} p<0.05,^{***} p<0.01$.}}
\end{table}
